The prognosis of patients with lung adenocarcinoma (LUAD), especially early-stage LUAD, is dependent on clinicopathological features. However, its predictive utility is limited. In this study, we developed and trained a DeepRePath model based on a deep convolutional neural network (CNN) using multi-scale pathology images to predict the prognosis of patients with early-stage LUAD. DeepRePath was pre-trained with 1067 hematoxylin and eosin-stained whole-slide images of LUAD from the Cancer Genome Atlas. DeepRePath was further trained and validated using two separate CNNs and multi-scale pathology images of 393 resected lung cancer specimens from patients with stage I and II LUAD. Of the 393 patients, 95 patients developed recurrence after surgical resection. The DeepRePath model showed average area under the curve (AUC) scores of 0.77 and 0.76 in cohort I and cohort II (external validation set), respectively. Owing to low performance, DeepRePath cannot be used as an automated tool in a clinical setting. When gradient-weighted class activation mapping was used, DeepRePath indicated the association between atypical nuclei, discohesive tumor cells, and tumor necrosis in pathology images showing recurrence. Despite the limitations associated with a relatively small number of patients, the DeepRePath model based on CNNs with transfer learning could predict recurrence after the curative resection of early-stage LUAD using multi-scale pathology images.

Background: A high platelet–lymphocyte ratio (PLR) is a marker of systemic inflammation and, together with the neutrophil–lymphocyte ratio (NLR), is associated with poor outcomes in several cancers. We investigated the prognostic value of PLR and other systemic inflammatory markers, such as NLR, systemic immune-inflammation index (SII), and systemic inflammation response index (SIRI), in oral squamous cell carcinoma (OSCC) patients undergoing surgical resection. Methods: We derived PLR, NLR, SII, and SIRI from a retrospective chart review of 269 consecutive OSCC patients. The complete blood count examined in the immediate preoperative period was used to compute PLR, NLR, SII, and SIRI. We analyzed the relationship between these systemic inflammatory markers and the clinicopathologic characteristics, disease-specific survival (DSS), and progression-free survival (PFS) of patients. Results: In the univariate analysis, high PLR and SII were significantly associated with worse DSS and PFS (all p < 0.05). In the multivariate analysis, PLR (HR 2.36, 95% CI 1.28–4.36 for DSS; HR 1.80, 95% CI 1.06–3.06 for PFS) was an independent predictor of survival outcomes. When PLR was analyzed as a continuous variable, the relationship between the outcome and preoperative PLR was not monotonically linear. In the subgroup analysis, PLR was more strongly associated with DSS and PFS in patients who were male, had stage III/IV OSCC, or had lymph node metastasis. Conclusion: Our data suggest that in OSCC patients, the pretreatment PLR is an independent predictor of DSS and PFS. The PLR is a readily available biomarker that will improve prognostication and risk stratification in OSCC.

Background: Invasive mucinous adenocarcinoma (IMA) of the lung is a rare and distinct subtype of adenocarcinoma that can appear as airspace opacities on computed tomography (CT). In daily practice, we have occasionally encountered spontaneous regression of airspace opacities (SRAs) without treatment on serial CTs in patients with IMAs, which has not previously been described in the literature. Here, we describe serial CT findings with emphasis on SRAs in relation to clinicopathological features and treatment outcomes in patients with IMAs.Methods: A total of 46 patients with pathologically‐confirmed IMAs of the lung from January 2013 to June 2018 were included. Serial CT scans were reviewed and the patients were classified into SRA and no‐SRA groups according to the presence of SRA. Radiological features, clinicopathological characteristics, and treatment outcomes were compared between the SRA and no‐SRA groups.Results: A total of 32 patients were included in the no‐SRA group and 14 patients in the SRA group. IMAs in the SRA group were mostly pneumonic (P < 0.001), larger (P < 0.001), multifocal (P = 0.001), and showed higher stage (P < 0.001) on initial CT. Of seven patients who died during follow‐up, six were from the SRA group (P < 0.001). Mean overall survival for all IMAs was 86.6 months (range, 0–110 months), and the SRA group showed significantly worse overall survival (P < 0.001).Conclusions: IMAs of the lung showing SRAs on serial CTs are larger and multifocal, and tend to be pneumonic in type on initial CT. Patients present at a higher stage of disease, with higher mortality rate and reduced overall survival.Key points: Significant findings of the study: Invasive mucinous adenocarcinomas (IMAs) of the lung can show spontaneous regression of airspace opacities (SRAs) on serial CTs, without being correlated to the administration of anticancer drugs. IMAs that showed SRAs demonstrated reduced overall survival in patients. What this study adds: When airspace opacities show regression on CT, IMA should still be included in the differential diagnosis. A more careful application of RECIST 1.1 is needed in the assessment of tumor response of IMAs.

Squamous cell carcinomas (SqCCs) of the lung are known to arise more often in a central area but reports of peripheral SqCCs have increased, with a pathogenesis that is obscured. In this study, the clinicopathologic characteristics of peripheral lung SqCCs were studied and compared with those of the central type. This study included 63 peripheral lung SqCCs and 48 randomly selected central cases; hematoxylin and eosin-stained slides of surgically resected specimens were reviewed in conjunction with radiologic images and clinical history. Cytokeratin-7 immunohistochemical staining of key slides and epidermal growth factor receptor (EGFR)/KRAS mutations tested by peptide nucleic acid clamping were also included. Stages of peripheral SqCCs were significantly lower than central SqCCs (p=.016). Cystic change of the mass (p=.007), presence of interstitial fibrosis (p=0.007), and anthracosis (p=.049) in the background lung were significantly associated with the peripheral type. Cytokeratin-7 positivity was also higher in peripheral SqCCs with cutoffs of both 10% and 50% (p=.011). Pathogenic mutations in EGFR and KRAS were observed in only one case out of the 72 evaluated. The Cox proportional hazard model indicated a significantly better disease-free survival (p=.009) and the tendency of better overall survival (p=.106) in the peripheral type. In peripheral type, lower stage is a favorable factor for survival but more frequent interstitial fibrosis and older age are unfavorable factors. Multivariate Cox analysis revealed that peripheral type is associated with better disease-free survival. The pathogenesis of peripheral lung SqCCs needs further investigation, together with consideration of the background lung conditions.

To assess the additive prognostic value of MR-based radiomics in predicting progression-free survival (PFS) in patients with nasopharyngeal carcinoma (NPC) Patients newly diagnosed with non-metastatic NPC between June 2006 and October 2019 were retrospectively included and randomly grouped into training and test cohorts (7:3 ratio). Radiomic features (n=213) were extracted from T2-weighted and contrast-enhanced T1-weighted MRI. The patients were staged according to the 8^th edition of American Joint Committee on Cancer Staging Manual. The least absolute shrinkage and selection operator was used to select the relevant radiomic features. Univariate and multivariate Cox proportional hazards analyses were conducted for PFS, yielding three different survival models (clinical, stage, and radiomic). The integrated time-dependent area under the curve (iAUC) for PFS was calculated and compared among different combinations of survival models, and the analysis of variance was used to compare the survival models. The prognostic performance of all models was validated using a test set with integrated Brier scores. This study included 81 patients (training cohort=57; test cohort=24), and the mean PFS was 57.5 ± 43.6 months. In the training cohort, the prognostic performances of survival models improved significantly with the addition of radiomics to the clinical (iAUC, 0.72–0.80; p=0.04), stage (iAUC, 0.70–0.79; p=0.001), and combined models (iAUC, 0.76–0.81; p<0.001). In the test cohort, the radiomics and combined survival models were robustly validated for their ability to predict PFS. Integration of MR-based radiomic features with clinical and stage variables improved the prediction PFS in patients diagnosed with NPC.

Sung YE, et al. J Gynecol Oncol. 2016 Jul;27(6):e56. https://doi.org/10.3802/jgo.2016.27.e56